skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Searching for pulsars using image pattern recognition

Abstract

In the modern era of big data, many fields of astronomy are generating huge volumes of data, the analysis of which can sometimes be the limiting factor in research. Fortunately, computer scientists have developed powerful data-mining techniques that can be applied to various fields. In this paper, we present a novel artificial intelligence (AI) program that identifies pulsars from recent surveys by using image pattern recognition with deep neural nets—the PICS (Pulsar Image-based Classification System) AI. The AI mimics human experts and distinguishes pulsars from noise and interference by looking for patterns from candidate plots. Different from other pulsar selection programs that search for expected patterns, the PICS AI is taught the salient features of different pulsars from a set of human-labeled candidates through machine learning. The training candidates are collected from the Pulsar Arecibo L-band Feed Array (PALFA) survey. The information from each pulsar candidate is synthesized in four diagnostic plots, which consist of image data with up to thousands of pixels. The AI takes these data from each candidate as its input and uses thousands of such candidates to train its ∼9000 neurons. The deep neural networks in this AI system grant it superior ability to recognize variousmore » types of pulsars as well as their harmonic signals. The trained AI's performance has been validated with a large set of candidates from a different pulsar survey, the Green Bank North Celestial Cap survey. In this completely independent test, the PICS ranked 264 out of 277 pulsar-related candidates, including all 56 previously known pulsars and 208 of their harmonics, in the top 961 (1%) of 90,008 test candidates, missing only 13 harmonics. The first non-pulsar candidate appears at rank 187, following 45 pulsars and 141 harmonics. In other words, 100% of the pulsars were ranked in the top 1% of all candidates, while 80% were ranked higher than any noise or interference. The performance of this system can be improved over time as more training data are accumulated. This AI system has been integrated into the PALFA survey pipeline and has discovered six new pulsars to date.« less

Authors:
; ; ; ;  [1];  [2];  [3]; ;  [4]; ; ; ; ; ;  [5];  [6]; ; ; ;
  1. Department of Physics and Astronomy, 6224 Agricultural Road, University of British Columbia, Vancouver, BC, V6T 1Z1 (Canada)
  2. Astronomy Department, Cornell University, Ithaca, NY 14853 (United States)
  3. Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, D-53121 Bonn (Germany)
  4. Department of Physics, McGill University, Montreal, QC H3A 2T8 (Canada)
  5. Center for Advanced Radio Astronomy, University of Texas at Brownsville, Brownsville, TX 78520 (United States)
  6. NRAO, Charlottesville, VA 22903 (United States)
Publication Date:
OSTI Identifier:
22348078
Resource Type:
Journal Article
Journal Name:
Astrophysical Journal
Additional Journal Information:
Journal Volume: 781; Journal Issue: 2; Other Information: Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0004-637X
Country of Publication:
United States
Language:
English
Subject:
79 ASTROPHYSICS, COSMOLOGY AND ASTRONOMY; ARTIFICIAL INTELLIGENCE; ASTRONOMY; CLASSIFICATION; DATA ANALYSIS; GALAXIES; HARMONICS; IMAGE PROCESSING; IMAGES; INTERFERENCE; NEURAL NETWORKS; NEUTRONS; NOISE; PATTERN RECOGNITION; PULSARS; STARS; TRAINING

Citation Formats

Zhu, W. W., Berndsen, A., Madsen, E. C., Tan, M., Stairs, I. H., Brazier, A., Lazarus, P., Lynch, R., Scholz, P., Stovall, K., Cohen, S., Dartez, L. P., Lunsford, G., Martinez, J. G., Mata, A., Ransom, S. M., Banaszak, S., Biwer, C. M., Flanigan, J., Rohr, M., E-mail: zhuww@phas.ubc.ca, E-mail: berndsen@phas.ubc.ca, and others, and. Searching for pulsars using image pattern recognition. United States: N. p., 2014. Web. doi:10.1088/0004-637X/781/2/117.
Zhu, W. W., Berndsen, A., Madsen, E. C., Tan, M., Stairs, I. H., Brazier, A., Lazarus, P., Lynch, R., Scholz, P., Stovall, K., Cohen, S., Dartez, L. P., Lunsford, G., Martinez, J. G., Mata, A., Ransom, S. M., Banaszak, S., Biwer, C. M., Flanigan, J., Rohr, M., E-mail: zhuww@phas.ubc.ca, E-mail: berndsen@phas.ubc.ca, & others, and. Searching for pulsars using image pattern recognition. United States. https://doi.org/10.1088/0004-637X/781/2/117
Zhu, W. W., Berndsen, A., Madsen, E. C., Tan, M., Stairs, I. H., Brazier, A., Lazarus, P., Lynch, R., Scholz, P., Stovall, K., Cohen, S., Dartez, L. P., Lunsford, G., Martinez, J. G., Mata, A., Ransom, S. M., Banaszak, S., Biwer, C. M., Flanigan, J., Rohr, M., E-mail: zhuww@phas.ubc.ca, E-mail: berndsen@phas.ubc.ca, and others, and. 2014. "Searching for pulsars using image pattern recognition". United States. https://doi.org/10.1088/0004-637X/781/2/117.
@article{osti_22348078,
title = {Searching for pulsars using image pattern recognition},
author = {Zhu, W. W. and Berndsen, A. and Madsen, E. C. and Tan, M. and Stairs, I. H. and Brazier, A. and Lazarus, P. and Lynch, R. and Scholz, P. and Stovall, K. and Cohen, S. and Dartez, L. P. and Lunsford, G. and Martinez, J. G. and Mata, A. and Ransom, S. M. and Banaszak, S. and Biwer, C. M. and Flanigan, J. and Rohr, M., E-mail: zhuww@phas.ubc.ca, E-mail: berndsen@phas.ubc.ca and others, and},
abstractNote = {In the modern era of big data, many fields of astronomy are generating huge volumes of data, the analysis of which can sometimes be the limiting factor in research. Fortunately, computer scientists have developed powerful data-mining techniques that can be applied to various fields. In this paper, we present a novel artificial intelligence (AI) program that identifies pulsars from recent surveys by using image pattern recognition with deep neural nets—the PICS (Pulsar Image-based Classification System) AI. The AI mimics human experts and distinguishes pulsars from noise and interference by looking for patterns from candidate plots. Different from other pulsar selection programs that search for expected patterns, the PICS AI is taught the salient features of different pulsars from a set of human-labeled candidates through machine learning. The training candidates are collected from the Pulsar Arecibo L-band Feed Array (PALFA) survey. The information from each pulsar candidate is synthesized in four diagnostic plots, which consist of image data with up to thousands of pixels. The AI takes these data from each candidate as its input and uses thousands of such candidates to train its ∼9000 neurons. The deep neural networks in this AI system grant it superior ability to recognize various types of pulsars as well as their harmonic signals. The trained AI's performance has been validated with a large set of candidates from a different pulsar survey, the Green Bank North Celestial Cap survey. In this completely independent test, the PICS ranked 264 out of 277 pulsar-related candidates, including all 56 previously known pulsars and 208 of their harmonics, in the top 961 (1%) of 90,008 test candidates, missing only 13 harmonics. The first non-pulsar candidate appears at rank 187, following 45 pulsars and 141 harmonics. In other words, 100% of the pulsars were ranked in the top 1% of all candidates, while 80% were ranked higher than any noise or interference. The performance of this system can be improved over time as more training data are accumulated. This AI system has been integrated into the PALFA survey pipeline and has discovered six new pulsars to date.},
doi = {10.1088/0004-637X/781/2/117},
url = {https://www.osti.gov/biblio/22348078}, journal = {Astrophysical Journal},
issn = {0004-637X},
number = 2,
volume = 781,
place = {United States},
year = {Sat Feb 01 00:00:00 EST 2014},
month = {Sat Feb 01 00:00:00 EST 2014}
}